The urgency for the design of novel chemotherapeutic agents for the treatment of malaria has been renewed in recent times due to the evolution of human malarial parasites, primarily Plasmodium falciparum, which are resistant to traditional drugs. Research into a vaccine seems a very plausible alternative, but after years of investigation, no clinically acceptable product has come to date. At the same time, there is also an increasing decline in the efficacy of insecticides against mosquito vectors. At present, more than two-thirds of the world's population--approximately 500 million people--are thought to live in malaria areas (Miller, 1989). It ranks eighth in the World Health Organization's (WHO) list of ten most prevalent diseases of the world (270 million infections a year) and ranks ninth of the ten most deadly diseases, claiming over 2 million lives a year (Cox, 1991; Marshall, 1991). Though chiefly confined to poor nations, there are recent reports of infections in the United States (Marshall, 1991) and Australia (Johnson, 1991), and ever increasing cases of travellers' malaria (Steffen and Behrens, 1992).
Comparative biochemical studies between the malaria parasite, P. falciparum and its host have revealed differences in a number of metabolic pathways. One such distinction is that the parasite relies exclusively on pyrimidine synthesis de novo because of its inability to salvage preformed pyrimidines (Sherman, 1979). Moreover, the mature human red blood cell has no recognised requirement for pyrimidine nucleotides (Gero and O'Sullivan, 1990). Major efforts have been directed towards the development of inhibitors of the pyrimidine biosynthetic pathway (Hammond et al., 1985; Scott et al., 1986; Prapunwattana et al., 1988; Queen et al., 1990; Krungkrai et al., 1992), confirming its potential as a chemotherapeutic locus. Current research into the molecular biology of the key pyrimidine enzymes is envisioned as a powerful tool, not only to get a better understanding of the parasite's biochemistry, but also to explore specific differences between the parasite and the mammalian enzymes.
Glutamine-dependent carbamoyl phosphate synthetase (CPSU, EC 6.3.5.5) catalyses the first committed and rate-limiting step in the de novo pyrimidine biosynthetic pathway of eukaryotic organisms (Jones, 1980). Moreover, because it catalyzes a complex reaction involving three catalytic units and several substrates and intermediates, it is a very interesting enzyme to study from a biochemical point of view. The structural relationship of CPSII to other pyrimidine enzymes varies in different organisms, making it a good subject for evolutionary studies.
The paucity of material that can be obtained from malarial cultures has hampered the isolation of adequate amounts of pure protein for analysis. The difficulty in purifying CPS is further augmented by its inherent instability. Studies using crude extracts from P. berghei (a rodent malaria) revealed a high molecular weight protein containing CPS activity, which was assumed to be associated with ATCase (Hill et al., 1981), a situation also found in yeast (Makoff and Radford, 1978). However, recent analysis by Krungkrai and co-workers (1990) detected separate CPSII and ATCase activities in P. berghei. Although CPS activity has been detected in P. falciparum (Reyes et al., 1982) until this current study there is no indication of its size nor its linkage with other enzymes in the pathway.
The glutamine-dependent activity of CPSII can be divided into two steps: (1) a glutaminase (GLNase) reaction which hydrolyzes glutamine (Gln) and transfers ammonia to the site of the carbamoyl phosphate synthetase; and (2) a synthetase reaction. where carbamoyl phosphate is synthesised from two molecules of adenosine triphosphate (ATP), bicarbonate and ammonia. The second activity involves three partial reactions: (a) the activation of bicarbonate by ATP; (b) the reaction of the activated species carboxyphosphate with ammonia to form carbamate; and (c) the ATP-dependent phosphorylation of carbamate to form carbamoyl phosphate (powers and Meister, 1978). Hence, there are two major domains in CPSII, the glutamine amidotransferase domain (GAT) and the carbamoyl phosphate synthetase domain (CPS) or simply synthetase domain. The glutaminase domain (GLNase) is a subdomain of GAT, while there are two ATP-binding subdomains in the synthetase domain.
In view of the similarities between the glutamine amidotransferase domain of CPS and other amidotransferases, it has been proposed that these subunits arose by divergent evolution from a common ancestral gene (=20 kDa) representing the GLNase domain and that particular evolution of the CPS GAT domain (=42 kDa which includes the putative structural domain only present in CPS) must have involved fusions and/or insertions of other sequences (Werner et al., 1935). The GAT of mammalian CPSI gene has been proposed to be formed by a simple gene fusion event at the 5' end of this ancestral gene with an unknown gene (Nyunoya et al., 1985).
The genes for the larger synthetase domains of various organisms were postulated to have undergone a gene duplication of an ancestral kinase gene resulting in a polypeptide with two homologous halves (Simmer et al., 1990). Unlike the subunit structure of E. coli and arginine-specific CPS of yeast, a further fusion of the genes encoding GAT and the synthetase domains was suggested to have formed the single gene specific for pyrimidine biosynthesis in higher eukaryotes. Conversely, Simmer and colleagues (1990) proposed that the arginine-specific CPS's (like cpa1 and cpa2 in yeast) as well as rat mitochondrial CPSI arose by defusion from the pyrimidine chimera.